In article <[EMAIL PROTECTED]>,
Radford Neal <[EMAIL PROTECTED]> wrote:
>In article <w0E0a.19826$[EMAIL PROTECTED]>,
>john v verkuilen <[EMAIL PROTECTED]> wrote:

<>"I am told by my physician that I need a serious operation and have been 
<>informed that there has not been a fatal outcome in the 20 operations carried
<>out by the physician.  Does this information give me an estimate of the 
<>potential postoperative mortality?"  The answer is yes, you have about a 
<>15% chance of postoperative death (given no other information).  

>This is absurd.  I think it must take years of exposure to frequentist
>statistics before one can accept the argument offered below for this
>statement, in defiance of all common sense.  (Of course, the argument
>is invalid even in terms of frequentist statistics, but it's the sort
>of fallacy that the frequentist viewpoint encourages.)

Unfortunately, one typical service course is enough to 
cause this indoctrination to the alpha-level religion.

<>Assuming no observed events, a 95% upper bound on the rate of occurence in n 
<>trials is 3/n.  It's based on the Poisson distribution and the argument is 
<>summarized on page 50 of the book cited.  (I can type it in if you want.)  

>----------------------------------------------------------------------------
>Radford M. Neal                                       [EMAIL PROTECTED]
>Dept. of Statistics and Dept. of Computer Science [EMAIL PROTECTED]
>University of Toronto                     http://www.cs.utoronto.ca/~radford
>----------------------------------------------------------------------------


-- 
This address is for information only.  I do not claim that these views
are those of the Statistics Department or of Purdue University.
Herman Rubin, Deptartment of Statistics, Purdue University
[EMAIL PROTECTED]         Phone: (765)494-6054   FAX: (765)494-0558
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to